© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 11-RES-JB
    11. 9983
    Prev
    Next
    INDEX
    Explanations

    Keith Haring was an American artist and social activist known for his street-inspired artwork. His pieces often communicated themes of birth, death, war, and the AIDS epidemic, using bold lines and bright colors to convey powerful messages in a simple yet profound manner. Haring's work was deeply influenced by his

    oai_token-act-pair · gpt-3.5-turbo

    references to situations or actions taking place in front of an audience or crowd

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    among
    -0.78
    berman
    -0.70
    leep
    -0.66
    unker
    -0.64
    pring
    -0.63
     rede
    -0.61
    between
    -0.60
    bach
    -0.60
    isure
    -0.59
    ennis
    -0.59
    POSITIVE LOGITS
     cameras
    1.23
     camera
    1.08
     crowds
    0.93
     mirror
    0.92
     microphones
    0.92
     audiences
    0.88
     mirrors
    0.87
     roaring
    0.87
     Camera
    0.86
     cheering
    0.85
    Activations Density 0.124%

    No Known Activations